My domain careerretirementcalculator.com went unreachable (ERR_CONNECTION_RESET) today, right after publishing a new post through an automated webhook that fires a Deploy Hook. Site itself is fine — this is a DNS-level issue.

Setup:

Domain registered through Vercel, nameservers are ns1.vercel-dns.com / ns2.vercel-dns.com Project Domains page shows "Valid Configuration" for the apex DNS Records panel looks completely normal — the only apex-related entries are the Vercel-managed ALIAS records (e8b2b3cf82d68886.vercel-dns-017.com and the * wildcard), no unexpected A record anywhere SSL certs are valid and unexpired The actual problem: Querying ns1.vercel-dns.com (198.51.44.13) directly for an A record on the apex returns 18.204.152.241 — not a Vercel IP. Response headers from that IP show Via: 1.0 heroku-router and X-Powered-By: Express, i.e. it's routing somewhere completely unrelated to my project.

This is consistent across every public resolver I've tried: Google (8.8.8.8), Cloudflare (1.1.1.1), Quad9 (9.9.9.9), and OpenDNS (208.67.222.222) — so it's not regional propagation or local caching.

For comparison, www.careerretirementcalculator.com resolves correctly to 216.198.79.1 / 64.29.17.1 and works fine.

Vercel's status page shows all systems operational, so this looks isolated to my domain's zone specifically rather than a broader incident.

Has anyone seen this before, or is there a way to get Vercel's DNS backend to re-check/reconcile this record? I'm on the Hobby plan so ticket-based DNS support isn't available to me — any pointers appreciated.

Edit: Update to my earlier report: as of 5:35pm on July 27, 2026, this has gotten worse. Originally only the apex domain (careerretirementcalculator.com) was affected; now www.careerretirementcalculator.com has also started resolving to the same wrong IP. This suggests the problem is spreading within my domain's zone, not a single stuck record.